Tuning was done with Bayesian optimisation with the following parameters:
Acquisition function: Expected Improvement
alpha: 0.05
xi: 1e-4
TC: 60+0.6
Number of iterations: 100
Initial points: 5
Batch size: 20 games
STC
http://tests.stockfishchess.org/tests/view/
5dee291e3cff9a249bb9e470
LLR: 2.97 (-2.94,2.94) [-1.50,4.50]
Total: 19586 W: 4382 L: 4214 D: 10990
LTC
http://tests.stockfishchess.org/tests/view/
5dee4e273cff9a249bb9e473
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 38840 W: 6315 L: 6036 D: 26489
Bench:
5033242
constexpr uint64_t ttHitAverageResolution = 1024;
// Razor and futility margins
constexpr uint64_t ttHitAverageResolution = 1024;
// Razor and futility margins
- constexpr int RazorMargin = 661;
+ constexpr int RazorMargin = 594;
Value futility_margin(Depth d, bool improving) {
Value futility_margin(Depth d, bool improving) {
- return Value(198 * (d - improving));
+ return Value(232 * (d - improving));
}
// Reductions lookup table, initialized at startup
}
// Reductions lookup table, initialized at startup